We sought to quantify the impact of the transition in household diets in Bangladesh on the reduction of stunting in children from six months to two years of age over the 13-year period from 1992 to 2005.
This study utilizes data from four rounds of two linked survey systems: the Bangladesh Household [Income and] Expenditure Surveys (H[I]ES) and the Child [and Mother] Nutrition survey (C[M]NS). The households included in the C[M]NS were a subsample of those contained in the H[I]ES. Across all years, we analyzed 3,484 children who were six to twenty-three months of age at the time of C[M]NS. Dietary change was measured through dietary patterns previously derived using principal component analysis. We used Blinder-Oaxaca decomposition to assess the impact of changing diets, other socioeconomic characteristics of households, and child level characteristics on child nutritional status over time.
For six dietary patterns, the prevalence of adherence changed over time. In a fully adjusted model, only the “modern” dietary pattern – characterized by processed soybean/palm oil, poultry and beef, beverages, generally more diversity, and less rice – was significantly associated with changes in child nutritional status. Indicators included in the decomposition models explained 64% of the observed change in length-for-age z-score (LAZ), 84% of the observed change in weight-for-age z-score (WAZ), and 180% of the observed change in weight-for-length z-scores (WFL). The dietary transition towards the “modern” diet over time was a significant predictor of better child LAZ and WAZ, but not quite significant for WFL (p-values: 0.005, 0.005, and 0.07, respectively). Other significant predictors of LAZ and WAZ included changes in household access to improved latrines and increased caregiver education. Increased household real income was not a significant predictor of improved nutritional status.
Conclusions : Changes in household characteristics, namely more diverse diets and improved latrines, along with increasing caregiver education, were able to largely explain the changes in child nutritional status over time. Increasing income, however, did not independently improve child growth. This points to the need to focus on nutrition-specific and -sensitive interventions to continue to decrease child malnutrition.
Funding Sources : This publication has been produced with the assistance of the European Union and the UK Department for International Development. The contents of this publication are the sole responsibility of its authors and do not necessarily reflect the views of the European Union or the UK Department for International Development